General Fusion (Canada)

The MAST-U Super-X divertor provides the opportunity to study fusion plasma exhaust under novel conditions. However, in order to study these conditions, advanced diagnostics are required. Following the development of the MAST-U Multi-Wavelength Imaging (MWI) diagnostic, we present the installation of a multi-delay coherence imaging spectroscopy (CIS) system within the MAST-U MWI, along with modifications made to the MWI for effective operation. This diagnostic will measure either carbon ion flow velocities and temperatures or electron densities through Dγ emission. We have extended previously developed techniques for wavelength calibration to account for errors due to the misalignment of interferometer components. In addition, we have developed a comprehensive calibration procedure to account for the temperature dependence of the instrument's delays by fitting to a linearly modified version of the delay equation presented by Veiras etal. [Appl. Opt. 49(15), 2769 (2010)]. Together, these procedures reduce the cost and hardware complexity of implementing CIS instruments when compared to those that use in situ or tunable laser calibration systems, as calibrations can be generated to good accuracy using previously measured data.

This research investigates the potential application of generative language models, especially ChatGPT, in aviation safety analysis as a means to enhance the efficiency of safety analyses and accelerate the time it takes to process incident reports. In particular, ChatGPT was leveraged to generate incident synopses from narratives, which were subsequently compared with ground truth synopses from the Aviation Safety Reporting System (ASRS) dataset. The comparison was facilitated by using embeddings from Large Language Models (LLMs), with aeroBERT demonstrating the highest similarity due to its aerospace-specific fine-tuning. A positive correlation was observed between synopsis length and their cosine similarity. In a subsequent phase, human factor issues involved in incidents as identified by ChatGPT were compared to human factor issues identified by safety analysts. A concurrence rate of 61% was found, with ChatGPT demonstrating a cautious approach towards attributing human factor issues. Finally, the model was used to attribute incidents to relevant parties. As no dedicated ground truth column existed for this task, a manual evaluation was conducted. ChatGPT attributed the majority of incidents to the Flight Crew, ATC, Ground Personnel, and Maintenance. This study opens new avenues for leveraging AI in aviation safety analysis.

The spall strength of polydimethylsiloxane silicone oils has been studied using the planar impact of thin flyers to generate large transient negative pressures near the free surface of target samples. The liquids were contained within sealed capsules in which a 4-µm-thick aluminized Mylar diaphragm formed a free surface at the back of the sample. The liquid targets were impacted by PMMA flyers at velocities ranging from 130 to 700 m/s using a 64-mm-bore gas-gun, thus allowing for large variations in the strain rate and incident shock pressure. The peak tension in the liquid was determined by monitoring the free surface velocity using a photonic Doppler velocimetry (PDV) system. The paper focuses on the study of a system of silicone oils having vastly different viscosities (5 cSt to 1000 cSt), but otherwise similar liquid properties. The effect of viscosity on spall strength is compared to previously published data and models.

Abstract An accurate calculation of heat load on the divertor plates in a tokamak must take into account kinetic effects, present when the electron velocity distribution function (EVDF) in the plasma column departs from a thermal distribution. In this work PARASOL‐1D, a particle‐in‐cell code with a Monte‐Carlo binary collision model, is used to find and explain the electron heat flux‐limiting factor αe and the heat transmission coefficient (HTC) γe in the complex SOL. We develop and test two simple models for these kinetic factors that take as input the temperature of the SOL and the temperature of the electrons striking the divertor plates. Both models assume a piecewise EVDF, with a symmetric bulk electron population and a high‐energy tail of electrons moving only in the direction of the nearest diverter plate. The model EVDF are fit to the simulation‐derived EVDF by allowing the variables for the densities and temperatures in the bulk and tail to vary independently. The block model, which assumes a bulk population of infinite parallel temperature in the bulk, is found to reproduce the kinetic factors for a wide range of conditions in the complex SOL much better than both the classical and Gaussian models (© 2012 WILEY‐VCH Verlag GmbH & Co. KGaA, Weinheim)

Direct numerical simulations were conducted of a fully turbulent canonical nozzle/jet configuration. For all cases, the target Reynolds number, based on the jet velocity and diameter, was specified as 7670 and the jet Mach number and coflow Mach number were varied. The effect of the nozzle lip on the turbulent flow exiting the nozzle was investigated, with particular emphasis on whether Reynolds stress profiles at the nozzle exit could be collapsed with profiles in the fully developed region, and whether local behaviour in the vicinity of the nozzle exit could be predicted using asymptotic theory. The DNS data were also used to investigate the effect of varying Mach number and coflow on the mean flow and whether the various flow cases could be collapsed using similarity arguments.

We have developed two sources of high flux radiation: one, designated as SPX II, exploits the principle of spherical convergence of a strong shock wave in a noble gas to generate a plasma at the center of a sphere hot enough to emit broadband radiation from the UV to the soft x-ray region of the spectrum; the other is based on the principle of vacuum spark, whereby accelerated electrons in a strong electric field in vacuum bombard an aluminum anode surface, leading to K-alpha line emission (1.7 keV). Typical measured outputs from the SPX II are: 186 W in the visible range (3500 - 2800 angstrom); 754 W in the UV range (2800 - 1800 angstrom); 650 W in the deep UV (1800 - 600 angstrom), and 5 W in the soft x ray (15 - 10 angstrom) for an operating repetition rate of the machine of 1 Hz. Typical measured output from the vacuum spark source is 30 W in the K-alpha line (7 angstrom) for a few kilowatt machine. The above are all powers measured beyond the pertinent filter used to select the radiation of interest, such as beryllium, aluminum, or other filters.